mirror of
https://github.com/torvalds/linux.git
synced 2025-08-15 14:11:42 +02:00

-----BEGIN PGP SIGNATURE----- iHUEABYKAB0WIQRAhzRXHqcMeLMyaSiRxhvAZXjcogUCaINCgQAKCRCRxhvAZXjc os+nAP9LFHUwWO6EBzHJJGEVjJvvzsbzqeYrRFamYiMc5ulPJwD+KW4RIgJa/MWO pcYE40CacaekD8rFWwYUyszpgmv6ewc= =wCwp -----END PGP SIGNATURE----- Merge tag 'vfs-6.17-rc1.mmap_prepare' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs Pull mmap_prepare updates from Christian Brauner: "Last cycle we introduce f_op->mmap_prepare() inc84bf6dd2b
("mm: introduce new .mmap_prepare() file callback"). This is preferred to the existing f_op->mmap() hook as it does require a VMA to be established yet, thus allowing the mmap logic to invoke this hook far, far earlier, prior to inserting a VMA into the virtual address space, or performing any other heavy handed operations. This allows for much simpler unwinding on error, and for there to be a single attempt at merging a VMA rather than having to possibly reattempt a merge based on potentially altered VMA state. Far more importantly, it prevents inappropriate manipulation of incompletely initialised VMA state, which is something that has been the cause of bugs and complexity in the past. The intent is to gradually deprecate f_op->mmap, and in that vein this series coverts the majority of file systems to using f_op->mmap_prepare. Prerequisite steps are taken - firstly ensuring all checks for mmap capabilities use the file_has_valid_mmap_hooks() helper rather than directly checking for f_op->mmap (which is now not a valid check) and secondly updating daxdev_mapping_supported() to not require a VMA parameter to allow ext4 and xfs to be converted. Commitbb666b7c27
("mm: add mmap_prepare() compatibility layer for nested file systems") handles the nasty edge-case of nested file systems like overlayfs, which introduces a compatibility shim to allow f_op->mmap_prepare() to be invoked from an f_op->mmap() callback. This allows for nested filesystems to continue to function correctly with all file systems regardless of which callback is used. Once we finally convert all file systems, this shim can be removed. As a result, ecryptfs, fuse, and overlayfs remain unaltered so they can nest all other file systems. We additionally do not update resctl - as this requires an update to remap_pfn_range() (or an alternative to it) which we defer to a later series, equally we do not update cramfs which needs a mixed mapping insertion with the same issue, nor do we update procfs, hugetlbfs, syfs or kernfs all of which require VMAs for internal state and hooks. We shall return to all of these later" * tag 'vfs-6.17-rc1.mmap_prepare' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: doc: update porting, vfs documentation to describe mmap_prepare() fs: replace mmap hook with .mmap_prepare for simple mappings fs: convert most other generic_file_*mmap() users to .mmap_prepare() fs: convert simple use of generic_file_*_mmap() to .mmap_prepare() mm/filemap: introduce generic_file_*_mmap_prepare() helpers fs/xfs: transition from deprecated .mmap hook to .mmap_prepare fs/ext4: transition from deprecated .mmap hook to .mmap_prepare fs/dax: make it possible to check dev dax support without a VMA fs: consistently use can_mmap_file() helper mm/nommu: use file_has_valid_mmap_hooks() helper mm: rename call_mmap/mmap_prepare to vfs_mmap/mmap_prepare
203 lines
5 KiB
C
203 lines
5 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* fs/bfs/file.c
|
|
* BFS file operations.
|
|
* Copyright (C) 1999-2018 Tigran Aivazian <aivazian.tigran@gmail.com>
|
|
*
|
|
* Make the file block allocation algorithm understand the size
|
|
* of the underlying block device.
|
|
* Copyright (C) 2007 Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
|
|
*
|
|
*/
|
|
|
|
#include <linux/fs.h>
|
|
#include <linux/mpage.h>
|
|
#include <linux/buffer_head.h>
|
|
#include "bfs.h"
|
|
|
|
#undef DEBUG
|
|
|
|
#ifdef DEBUG
|
|
#define dprintf(x...) printf(x)
|
|
#else
|
|
#define dprintf(x...)
|
|
#endif
|
|
|
|
const struct file_operations bfs_file_operations = {
|
|
.llseek = generic_file_llseek,
|
|
.read_iter = generic_file_read_iter,
|
|
.write_iter = generic_file_write_iter,
|
|
.mmap_prepare = generic_file_mmap_prepare,
|
|
.splice_read = filemap_splice_read,
|
|
};
|
|
|
|
static int bfs_move_block(unsigned long from, unsigned long to,
|
|
struct super_block *sb)
|
|
{
|
|
struct buffer_head *bh, *new;
|
|
|
|
bh = sb_bread(sb, from);
|
|
if (!bh)
|
|
return -EIO;
|
|
new = sb_getblk(sb, to);
|
|
memcpy(new->b_data, bh->b_data, bh->b_size);
|
|
mark_buffer_dirty(new);
|
|
bforget(bh);
|
|
brelse(new);
|
|
return 0;
|
|
}
|
|
|
|
static int bfs_move_blocks(struct super_block *sb, unsigned long start,
|
|
unsigned long end, unsigned long where)
|
|
{
|
|
unsigned long i;
|
|
|
|
dprintf("%08lx-%08lx->%08lx\n", start, end, where);
|
|
for (i = start; i <= end; i++)
|
|
if(bfs_move_block(i, where + i, sb)) {
|
|
dprintf("failed to move block %08lx -> %08lx\n", i,
|
|
where + i);
|
|
return -EIO;
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
static int bfs_get_block(struct inode *inode, sector_t block,
|
|
struct buffer_head *bh_result, int create)
|
|
{
|
|
unsigned long phys;
|
|
int err;
|
|
struct super_block *sb = inode->i_sb;
|
|
struct bfs_sb_info *info = BFS_SB(sb);
|
|
struct bfs_inode_info *bi = BFS_I(inode);
|
|
|
|
phys = bi->i_sblock + block;
|
|
if (!create) {
|
|
if (phys <= bi->i_eblock) {
|
|
dprintf("c=%d, b=%08lx, phys=%09lx (granted)\n",
|
|
create, (unsigned long)block, phys);
|
|
map_bh(bh_result, sb, phys);
|
|
}
|
|
return 0;
|
|
}
|
|
|
|
/*
|
|
* If the file is not empty and the requested block is within the
|
|
* range of blocks allocated for this file, we can grant it.
|
|
*/
|
|
if (bi->i_sblock && (phys <= bi->i_eblock)) {
|
|
dprintf("c=%d, b=%08lx, phys=%08lx (interim block granted)\n",
|
|
create, (unsigned long)block, phys);
|
|
map_bh(bh_result, sb, phys);
|
|
return 0;
|
|
}
|
|
|
|
/* The file will be extended, so let's see if there is enough space. */
|
|
if (phys >= info->si_blocks)
|
|
return -ENOSPC;
|
|
|
|
/* The rest has to be protected against itself. */
|
|
mutex_lock(&info->bfs_lock);
|
|
|
|
/*
|
|
* If the last data block for this file is the last allocated
|
|
* block, we can extend the file trivially, without moving it
|
|
* anywhere.
|
|
*/
|
|
if (bi->i_eblock == info->si_lf_eblk) {
|
|
dprintf("c=%d, b=%08lx, phys=%08lx (simple extension)\n",
|
|
create, (unsigned long)block, phys);
|
|
map_bh(bh_result, sb, phys);
|
|
info->si_freeb -= phys - bi->i_eblock;
|
|
info->si_lf_eblk = bi->i_eblock = phys;
|
|
mark_inode_dirty(inode);
|
|
err = 0;
|
|
goto out;
|
|
}
|
|
|
|
/* Ok, we have to move this entire file to the next free block. */
|
|
phys = info->si_lf_eblk + 1;
|
|
if (phys + block >= info->si_blocks) {
|
|
err = -ENOSPC;
|
|
goto out;
|
|
}
|
|
|
|
if (bi->i_sblock) {
|
|
err = bfs_move_blocks(inode->i_sb, bi->i_sblock,
|
|
bi->i_eblock, phys);
|
|
if (err) {
|
|
dprintf("failed to move ino=%08lx -> fs corruption\n",
|
|
inode->i_ino);
|
|
goto out;
|
|
}
|
|
} else
|
|
err = 0;
|
|
|
|
dprintf("c=%d, b=%08lx, phys=%08lx (moved)\n",
|
|
create, (unsigned long)block, phys);
|
|
bi->i_sblock = phys;
|
|
phys += block;
|
|
info->si_lf_eblk = bi->i_eblock = phys;
|
|
|
|
/*
|
|
* This assumes nothing can write the inode back while we are here
|
|
* and thus update inode->i_blocks! (XXX)
|
|
*/
|
|
info->si_freeb -= bi->i_eblock - bi->i_sblock + 1 - inode->i_blocks;
|
|
mark_inode_dirty(inode);
|
|
map_bh(bh_result, sb, phys);
|
|
out:
|
|
mutex_unlock(&info->bfs_lock);
|
|
return err;
|
|
}
|
|
|
|
static int bfs_writepages(struct address_space *mapping,
|
|
struct writeback_control *wbc)
|
|
{
|
|
return mpage_writepages(mapping, wbc, bfs_get_block);
|
|
}
|
|
|
|
static int bfs_read_folio(struct file *file, struct folio *folio)
|
|
{
|
|
return block_read_full_folio(folio, bfs_get_block);
|
|
}
|
|
|
|
static void bfs_write_failed(struct address_space *mapping, loff_t to)
|
|
{
|
|
struct inode *inode = mapping->host;
|
|
|
|
if (to > inode->i_size)
|
|
truncate_pagecache(inode, inode->i_size);
|
|
}
|
|
|
|
static int bfs_write_begin(const struct kiocb *iocb,
|
|
struct address_space *mapping,
|
|
loff_t pos, unsigned len,
|
|
struct folio **foliop, void **fsdata)
|
|
{
|
|
int ret;
|
|
|
|
ret = block_write_begin(mapping, pos, len, foliop, bfs_get_block);
|
|
if (unlikely(ret))
|
|
bfs_write_failed(mapping, pos + len);
|
|
|
|
return ret;
|
|
}
|
|
|
|
static sector_t bfs_bmap(struct address_space *mapping, sector_t block)
|
|
{
|
|
return generic_block_bmap(mapping, block, bfs_get_block);
|
|
}
|
|
|
|
const struct address_space_operations bfs_aops = {
|
|
.dirty_folio = block_dirty_folio,
|
|
.invalidate_folio = block_invalidate_folio,
|
|
.read_folio = bfs_read_folio,
|
|
.writepages = bfs_writepages,
|
|
.write_begin = bfs_write_begin,
|
|
.write_end = generic_write_end,
|
|
.migrate_folio = buffer_migrate_folio,
|
|
.bmap = bfs_bmap,
|
|
};
|
|
|
|
const struct inode_operations bfs_file_inops;
|